The OX40-receptor (OX40R) (also known as CD134, ACT-4, ACT35) is a member of the TNF receptor family which is expressed on activated CD4+ T-cells (see WO 95/12673). Triggering of this receptor via the OX40-ligand, named OX40L, gp34 or ACT-4-ligand, present on activated B-cells and dendritic cells, enhances the proliferation of CD4+ T-cells during an immune response and influences the formation of CD4+ memory T-cells. Furthermore, the OX40R-OX40L system mediates adhesion of activated T-cells to endothelial cells, thus directing the activated CD4+ T-cells to the site of inflammation.
Inflammatory and autoimmune diseases, such as rheumatoid arthritis and inflammatory bowel disease, are characterized by an infiltration of activated T-cells at the site of inflammation, which is believed to orchestrate the response leading to chronic tissue destruction. In patients with inflammatory bowel disease, OX40+ CD4+ T-cells can be found in the gut associated with sites of inflammation. In addition, in patients suffering from acute graft-vs-host-disease, elevated levels of OX40+ peripheral CD4+ T-cells are present in peripheral blood. In rheumatoid arthritis patients, OX40+ CD4+ T-cells are present in synovial fluid, while they are virtually absent from peripheral blood. Furthermore, OX40+ CD4+ T-cells are found in inflamed synovial tissue in addition to cells expressing the ligand for the OX40-receptor. This is in contrast to patients suffering from osteoarthritis, a joint disease that is not mediated by inflammation, where both cell types could not be found in significant numbers.
Thus, in patients suffering from several inflammatory disorders elevated levels of OX40+ CD4+ T-cells are present at sites of inflammation, indicating that these cells may be involved in progression of autoimmune disease. A blockade of the OX40R-OX40L pathway using antibodies or fusion proteins has led to the attenuation of disease progression in several animal models of autoimmune disease.
Besides their presence in autoimmune diseases, it has been shown that OX40+ T-cells are present within tumor lesions containing tumor infiltrating lymphocytes and in tumor cell positive draining lymph nodes (Weinberg et al., 2000). It was shown in several tumor models in mice that engagement of the OX40-receptor in vivo during tumor priming significantly delayed and prevented the appearance of tumors as compared to control treated mice (Weinberg et al., 2000). Hence, it has been contemplated to enhance the immune response of a mammal to an antigen by engaging the OX40-receptor by administering an OX40-receptor binding agent (WO 99/42585; Weinberg et al., 2000). One possibility is to use a natural ligand of the OX40-receptor, i.e. the OX40-ligand, or fusion proteins thereof as an OX40-receptor binding ligand. Such proteins however have a fixed affinity for the receptor that is not easily changed, may not have the circulatory retention time to exert the desired therapeutic effect, and may give rise to immunogenicity (Weinberg et al., 2000).
Another possibility to stimulate T-cells by virtue of the OX40-receptor pathway, is to use antibodies against this receptor (Kaleeba et al., 1998; Weinberg et al., 2000). A rat anti-mouse OX40-receptor antibody named OX86 (Al-Shamkhani et al., 1996) appeared to engage the OX40-receptor in murine tumor models (Weinberg et al., 2000; U.S. Pat. No. 6,312,700).
To our knowledge agonistic antibodies, particularly human agonistic antibodies, that are capable of stimulating the human OX40-receptor have not been disclosed in the art. Furthermore, it is well known that non-human antibodies are limited in their use in vivo in humans. Problems associated with administration of non-human antibodies to humans are inter alia short serum half life, an inability to trigger certain human effector functions and elicitation of an unwanted dramatic immune response against the non-human antibody in a human.
In general, attempts to overcome the problems associated with use of fully non-human antibodies in humans, have involved genetically engineering the antibodies to be more “human-like”. A first stage in the humanization process was preparing chimeric antibodies, i.e. antibodies in which the variable regions of the antibody chains are derived from the non-human species and the constant regions of the antibody chains are human-derived. Subsequently, domains between the variable domains which specify the antigen binding were replaced by their human counterparts leading to so-called humanized antibodies. A disadvantage of these chimeric and humanized antibodies is that they still retain some non-human sequences and therefore still elicit an unwanted immune reaction, especially when administered for prolonged periods.
In the light of the above, there is still a need for human antibodies that stimulate the human OX40-receptor. These antibodies can be useful in inter alia the treatment and/or prevention of tumours in humans.